>>>> "Hrvoje" == Hrvoje Niksic
<hniksic(a)srce.hr> writes:
"Stephen J. Turnbull" <turnbull(a)sk.tsukuba.ac.jp> writes:
> Making people who have little direct interest in MULE conform
> to complex coding conventions that they don't understand is
> bad. If functions that must be MULE-aware could be localized
> to a smaller set of modules, it would be a good thing.
Hrvoje> You can't localize general concepts like "characters are
Hrvoje> not 1 byte".
Of course you can, by making an opaque character object type and
manipulating instances only through messages. We may not like the
amount of work involved or the efficiency loss. The fact that 66 of
the 201 .c files in my current xemacs-21/src directory contain a
reference to Emchar or Bufbyte suggests it'd be horrendous. :-(
Hrvoje> If third-party contributors don't understand
Hrvoje> Mule coding conventions, too bad for them. Besides, the
Hrvoje> conventions are not all that complex, once you get to know
Hrvoje> them -- which goes for the rest of the XEmacs C sources.
<DIV ATTITUDE="tongue-in-cheek">
You are of course referring to the GCPRO convention?
</DIV>
I don't contest your points, pragmatically. But even now we still
have problems with integer vs. character comparisons cropping up every
once in a while.
Also, eventually we are going to have to modernize Mule; the current
implementation is just too constraining with the Japanese talking
about creating a completely new character set standard which will have
to coexist with the old one, only about 20% of the Chinese standards
implemented as yet, and Unicode not implemented at all. (Morioka-san
did send some patches for Unicode/UCS, which I haven't looked at
closely. There was no discussion at the time, and they haven't been
incorporated.)
If some of the underlying Mule implementation could be off-loaded into
a module with a specific set of calling conventions, it might help to
formalize the mess a bit, and head off some of the inevitable
confusion when that modernization happens.
--
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Institute of Policy and Planning Sciences Tel/fax: +81 (298) 53-5091
__________________________________________________________________________
__________________________________________________________________________
What are those two straight lines for? "Free software rules."